Level and correlates of physical activity among children and adolescents with juvenile idiopathic arthritis compared to controls: results from a German nationwide prospective observational cohort study

Juvenile idiopathic arthritis (JIA) is a heterogeneous group of inflammatory diseases of unknown origin, with onset before the age of 16 years. JIA is the most common chronic rheumatic disease affecting the pediatric population, with prevalence numbers ranging from 2 to 20 per 100,000 in European countries and up to 168 per 100,000 in Germany [1, 2]. Characterized by an overproduction of inflammatory cytokines at the joint level, JIA leads to a chronic inflammatory state accompanying joint pain, fatigue, stiffness and movement restrictions [3]. These symptoms and certain medical treatments for managing JIA place children at increased risk of sub-optimal bone mineralization and osteoporosis, malnutrition, muscle weakness, mobility impairments, and limitations in activities of daily living, such as playing [3‐7].

Promoting healthy lifestyles including physical activity (PA) is an intuitively attractive strategy to alleviate several disease-related symptoms such as low aerobic fitness, pain, fatigue, muscle weakness and poor health-related quality of life [5, 8‐10]. Although the propensity of young patients with JIA to be physically hypoactive can be detrimental to general disease symptoms and function [11], previous studies have shown that they tend to be less physically active [12, 13], spend more time in sedentary behaviour and less time in WHO [14] recommended health-enhancing PA of moderate-to-vigorous intensity compared to healthy controls [15].

However, previous data on PA levels among children and adolescents with JIA were frequently based on small samples, considered only certain JIA categories and age ranges, and did not provide data of representative controls, thus leading to inconsistent conclusions. There is currently a lack of reliable information on clinical barriers to PA in this population, knowledge of which, however, is a prerequisite for promoting active lifestyles and deriving appropriate interventions.

The aim of this study was to provide information on PA levels in a very large JIA cohort compared to representative population controls. We examined the extent of pronounced physical inactivity and its potential correlates in order to identify clinical barriers and risk groups for a physically inactive lifestyle.

For the year 2017, data from 6,297 age- and sex-matched pairs of 3–17 year olds could be considered. Approximately two-thirds of the sample were female. The number of pairs varied between age groups (Table 1).

The results of this nationwide representative study in children and adolescents with JIA extend the current literature by providing information on self-reported PA levels and their cross-sectional correlates. In 2017, patients with JIA achieved the WHO-recommended minimum level of PA comparable to or even more frequently than the general population, depending on JIA category. However, a significant decrease in PA level was observed with increasing age in both patients and controls.

At least eight in ten adolescents with JIA did not achieve the minimum level of physical activity, while one to two in ten (15%) were even considered highly physically inactive. The proportion of physical inactivity was comparable to general population controls. In patients, female sex, longer disease duration, older age at JIA onset, more functional limitations and higher disease activity were associated with physical inactivity.

Our data are based on one of the most extensive analyses of PA levels and its correlates conducted in pediatric JIA patients to date.

We found that the overall group of children and adolescents with JIA had higher PA levels on average than the German general population, although PA levels between patients and controls converged with age. Thus, our results differ from recent reports showing that young patients with JIA have lower PA levels than healthy peers [13, 25]. Other studies providing information on PA state that overall PA levels in JIA patients diagnosed in the era of biologics were similar to controls [15, 26]. In this context, Sherman et al. [26] postulate that this lack of difference may be due to clinical remission following an early, aggressive treat-to-target strategy. However, it should be noted that these studies reported PA levels only for the whole group, probably due to small sample sizes, narrow age ranges and underrepresentation of certain JIA categories (particularly oligoarthritis). In contrast, the distribution of categories in our cohort differs only slightly from that previously reported for Europe [27]. As we found significant variations in PA levels between JIA categories in our large, representative sample, this may partly explain discrepancies with overall PA reported in other studies. However, it is also conceivable that other characteristics not recorded in the study populations, such as socioeconomic status or cultural characteristics, may complicate comparability.

Patients with systemic arthritis and persistent oligoarthritis were significantly more likely to achieve the recommended PA level than age- and sex-matched controls. Consistent with previous studies, patients with enthesitis-related arthritis [28] and rheumatoid factor-positive polyarthritis [29] reported lowest PA levels. This could be explained on the one hand by a higher disease burden due to a higher number of active joints and on the other hand by the affection of the axial skeleton and a higher presence of enthesopathy with increasing pain in the lower extremities during all weight-bearing activities. In addition, the location of arthritis in the lower limbs, particularly in hip and knee joints, appears to be important as these affect walking endurance, cycling performance, function and quality of life [30]. Further, the onset of disease in enthesitis-related arthritis and rheumatoid factor-positive polyarthritis often occurs during puberty, an age phase generally associated with lower PA levels even in healthy individuals.

This age phase is usually accompanied by increasing independence from parents, changing life circumstances (e.g. employment) and many competing leisure activities (e.g. electronic media use) [31, 32]. Therefore, psychosocial and environmental changes are disease-independent factors that may have contributed to lower PA levels among adolescent patients and population controls.

Regardless, it should be noted that the WHO guidelines for PA of at least 60 min per day are only a minimum recommendation and that any PA beyond this may have additional health benefits [14]. This is reflected in the national guidelines for PA promotion in Germany, recommending at least 180 min of daily PA for children of kindergarten age and 90 min of daily PA for children and adolescents in school age [33].

In our study, one in ten children and adolescents aged 3 to 17 years was classified as highly physically inactive, with a comparable proportion in patients and controls. With slight sex differences to the disadvantage of girls, the proportion of physically inactive individuals increased during puberty, while the proportion of those who reached the recommended minimum level of PA decreased simultaneously (two out of ten adolescents). The observed decline in PA with age has already been described in several studies in healthy children and adolescents and is associated with an average decrease of 6 min per day per year of moderate to vigorous PA [34].

We identified as correlates of physical inactivity longer disease duration, older age at disease onset, higher disease activity and more functional limitations. Thus, our results are comparable to previous findings reporting associations between PA level, disease duration and disease activity [35, 36]. Although non-pharmacological, physical activity-based therapies are increasingly becoming part of the JIA treatment protocol [37], previous results have suggested that physical inactivity is not only the result of disease-related impairments [17, 25]. This observation can also be confirmed by the results within the framework of our study. Considering existing cut-off values [24], 51% of physically inactive adolescents (≥ 12 years) with oligoarthritis and 38% with polyarthritis were in a state of minimal disease activity. Although we cannot rule out that other clinical factors are responsible for physical inactivity in some patients, this finding highlights the importance of applying behavior change techniques to adolescents. As studies indicate that measures to contain the COVID-19 pandemic have had negative impacts on PA among European children and adolescents [38], this becomes all the more important. However, negative effects of pandemic-related restrictions were not reflected in the level of self-reported PA in our patients, at least until 2022. In patients with JIA, chronic inflammation with persisting systemic circulating inflammatory proteins constitute a risk for early vascular damage, with general physical inactivity further increasing this risk.

The strength of our study includes prospective cross-sectional evaluations using an observational cohort study with representative data on clinical characteristics as well as treatment assignments of children and adolescents with JIA in Germany. According to estimates, the number of JIA patients included corresponds to almost 45% of all expected JIA cases in Germany [39]. PA levels were assessed analogously to the methodology used in the reference population [19], considering clinically relevant parameters as well as general and disease-specific instruments.

Nevertheless, our results must be interpreted with caution given several possible limitations. We did not capture PA intensity and are therefore unable to accurately comment on whether this cohort of children and adolescents is meeting current PA recommendations of 60 min of moderate to vigorous PA per day. As respondents were not given any details or examples when completing the question on PA, it cannot be ruled out that PA was sometimes interpreted differently by individuals. Furthermore, the question on PA was based on self-reports. Although, most PA self-reports are suitable for classifying subjects according to their PA levels and are commonly used in pediatric and population-based research [40‐43], it is well recognized that their accuracy is limited and can lead to over- or under-reporting for a variety of reasons. In our JIA cohort, socially desirable responding in the presence of hospital staff, for example, might be one possible reason. Moreover, parent-reported outcomes may have led to an overestimation of PA levels in the younger cohort (patients aged < 12 years). Based on the parental education level recorded, most of our patients probably have a medium to high socioeconomic status (SES). Since the NPRD does not ask about income, we were not able to relate SES to PA. Finally, we were not able to examine health-related quality of life, which is known to be associated with PA levels in many patient and healthy populations.

In conclusion, children and adolescents with JIA are similarly or even more likely to achieve the WHO recommended minimum level of PA compared to general population controls, however, an overwhelming proportion is insufficiently physically active, partly despite satisfactory control of inflammation. As adolescent patients and patients with enthesitis-related arthritis, rheumatoid factor-positive polyarthritis as well as psoriatic arthritis are particularly vulnerable to physical inactivity, PA should be promoted specifically for these subgroups and for all patients with symptoms allowing PA. To achieve this, all those working with children and young people with JIA need to encourage them to engage in physical activities appropriate to their symptoms, lifestyle and fitness level.

In order to promote activities of daily life and to implement adequate interventions, JADAS and CHAQ should be controlled. To clarify both the safety of PA and the health risks associated with physical inactivity, efforts are needed to improve the quality of information provided for parents, health professionals, teachers and patients. Future studies should provide a better understanding of how socioeconomic status, health-related quality of life, and other patient-reported outcomes (e.g. mental health, sleep quality) relate to PA behaviour in JIA.